The Problem: Molten lead, used as a heating medium to pyrolyze plastics and rubber waste in the prior art, presents unique problems because lead is about 11.5 times heavier than the waste—the waste is quickly forced to the surface preventing contact time with the lead long enough to convert the waste in a reasonable amount of time. Particularly when solid waste includes polyolefins, poly(vinyl aromatic)s, and rubber from worn out tires, it is difficult to provide an economical level of conversion to reusable hydrocarbons within a residence time (in the molten bath) of less than 1 hour, preferably less than 30 min. “Reusable hydrocarbons” refers to both higher molecular weight hydrocarbons which are condensed, and lower molecular weight hydrocarbons which can be recycled as fuel. Reusable hydrocarbons consist of a major proportion by weight of condensible C5+ hydrocarbons (having at least five carbon atoms) and a minor proportion (relative to the C5+ hydrocarbons) of non-condensible C1-C4 hydrocarbons, typically less than 20% by weight of the C5+ hydrocarbons, the components in the vapor phase being in equilibrium with those in the condensate at the temperature and pressure conditions of condensation within the condenser.
Though a molten lead bath is able to provide a source of heat at a chosen, substantially constant temperature, using molten lead (or “melt”) as a heat transfer medium in a substantially oxygen-free atmosphere in the reactor, presents numerous difficulties. To begin with, a floating layer of organic waste acts as an insulating barrier, preventing pieces of waste within the floating layer from being heated sufficiently to depolymerize. If the waste cannot be adequately contacted with the melt it does not matter how much melt is in the bath. Yet, efficient heat transfer from the melt to the waste, to obtain an economic residence time in the melt, must not interfere with being able to transport the waste longitudinally through the melt. To cope with this problem by providing a high enough bath temperature to effect the pyrolysis in a reasonable amount of time, results in too high a production of hydrocarbons lower than C4, appreciable CO and CO2. To complicate the problem, when using a solid, particulate, catalyst it is critical that the waste be contacted and mixed with both the catalyst and the melt.
When such a catalyst is a combination of an aluminum powder and aluminum oxide mineral, whether calcined hydrated alumina, or calcined zeolite, this mixing is difficult to do without using a fluid bed. “Zeolite” refers to a natural or synthetic composition typically having the structure Mx/n[(AlO2)x(SiO2)y.zH2) where n is the charge of the metal cation, Mn+, which is usually Na+, K+, or Ca2+, x and y are integers, typically having substantially the same value in the range from 2 to 10, and the z is the number of moles of water of hydration.
Since conversion of scrap rubber generates sulfur and sulfur-containing compounds, the catalyst, most preferably a combination of aluminum powder and calcined bauxite powder, is required to be substantially unreactive with both, the sulfur and sulfur-containing compounds, and chlorine and hydrochloric (HCl) acid gases, if such gases are present in an appreciable amount. In addition, the reactor requires an essentially oxygen-free atmosphere within it; and the high specific gravity of lead precludes using very much of the melt in the bath, for practical cost considerations relating to the structural requirements of a vat or trough in which the molten lead bath is held.
Moreover, though the housing and other components of the reactor are typically made of acid and heat-resistant sheet steel, e.g. H25N20S2, the steel does not have notably long-term resistance to SO2, H2SO3, chlorine and HCl gases. The reliance on affordable steel and the use of aluminum powder in the catalyst requires feeding plastic substantially free of a halogen-containing synthetic resin, to the reaction zone, if safe, long-term operation of the reactor is sought. By “substantially free of a halogen-containing synthetic resin” is meant that less than 5% by weight of the waste is a polymer containing chlorine, bromine, iodine or fluorine, e.g. poly(vinyl chloride) (“PVC”) scrap, or other halogen-containing synthetic resins, e.g. chlorofluoro-, chlorobromo- and fluorocarbon polymers.
The Prior Art
Molten metal, particularly lead, has been the heat transfer medium of choice for the thermal conversion of organic matter, generally. The problem of heating organic matter which floated on a molten lead bath was recognized as early as before 1926 when U.S. Pat. No. 1,601,777 disclosed moving crushed shale along the undersurface of a slightly inclined apertured member, beneath the surface of a heated bath. U.S. Pat. No. 2,459,550 addresses the problem by confining wood or coal pieces between two endless screens. U.S. Pat. No. 3,977,960 teaches using angularly inclined screw conveyors to force crushed shale into a molten bath. As recently as 1990, U.S. Pat. No. 4,925,532 teaches moving perforated baskets filled with waste on an endless conveyor; the baskets are hooked to the conveyor to prevent them from floating against guide rails above the baskets. The '532 patent teaches that it is critical that the molten lead bath be maintained above 343° C. (650° F.), ignoring the fact that the melting point of pure lead at atmospheric pressure is just below, i.e. 327.5° C. (621.5° F.). It failed to realize that a catalyst could enhance conversion; and it missed the fact that optimum conversion of polyolefins, polystyrene and scrap from tires, to vapor consisting essentially of a major proportion by weight of C5+ hydrocarbons occurs only in the narrow range from 450° C.-550° C. (842° F.-1022° F.), a range commencing more than 100° C. above the temperature deemed critical. Most recently, in 1992, U.S. Pat. No. 5,085,738 teaches using a long, upwardly inclined oxygen-free cylindrical chamber filled with molten lead, through which chamber pieces of scrap tires are forced. A ram is used to circumvent the problem of floating rubber, but still relying solely on the thermal pyrolysis of the submerged rubber. The prior art countered the high specific gravity of molten lead by confining the charge in the melt. It ignores the problem of essentially instantly solidifying molten lead on the rubber as it is fed, because of the low heat capacity (and specific heat) of the lead; and, the requirement of timely supplying adequate heat to re-melt the lead.
It will be evident that the invention disclosed herebelow, for feeding the waste to the reactor, converting the waste in the reactor, removing and disposing of the residue, is based on the use of a unique catalyst in combination with a novel and unexpectedly efficient system of dealing with the numerous problems associated with feeding waste and catalyst to a molten lead bath in a sealed environment, including, for practical operation of the reactor, not submerging the waste in the molten lead. Further, not unexpectedly, the prior art processes and apparatus which rely solely on thermal pyrolysis of plastics and rubber in molten lead, are conspicuously devoid of data showing the effectiveness of the conversions obtained. As will be evident from the data presented below, the conversion of waste to reusable hydrocarbons by pyrolysis in molten lead alone, is only 53% (see Table 1) when the scrap is PE (polyethylene) and PP (polypropylene); and more than 90% when the catalyst used is bauxite/Al=97/3.
Recognizing the advantage of using an effective catalyst for the conversion of waste polyolefins, polystyrene and the like to hydrocarbons, U.S. Pat. No. 4,851,601 teaches using a fluid bed of zeolite particles, as does Chinese patent application WO95/06682. In each case, hydrocarbons having a wide range of boiling points are collected, but they rely on the efficient heat transfer provided by a fluid bed and the catalytic effect of a zeolite only, and the zeolite, by itself is evidently unaffected by the presence of chlorine in PVC.